Too Much Talking: why less in consultations may be more

I talk to much in consultations. Frequently I hear myself explaining to the patient: “well, that was a lot of me talking… any questions?” Inevitably there is, more often than not: “Can you just remind me what you said at the start again…” 

In my defence, over recent years there has been significant change in how policy and guideline makers would like us to consult, and it’s a case of more, not less. Take the recent NICE guideline on managing depression in adults. It feels like a hundred things need to be discussed, all before we’ve even mentioned anti-depressants and warning people about the negatives while trying to sell up on the positives, and, without so much as a sniff of an SSRI, how awful it will be to stop. 

Of course, discussing withdrawal symptoms is important and necessary, but this is another addition to our lengthy consultations. It has only been a handful of years since psychiatrists agreed these actually exist. In 2019 a systematic review published in Addictive Behaviours journal indicated more than half of people (56%) attempting to stop anti-depressants experienced withdrawal effects, with half of these classing them as severe. In 2019 the RCPsych changed its historical view on withdrawal symptoms in a position statement, reflected in the subsequent 2022 update to the NICE guideline on managing depression.

Such discussion takes time, always precious in general practice, but could all this talk and information actually be doing more harm than good? 

A new systematic review and meta-analysis of the incidence of antidepressant discontinuation symptoms has published in the Lancet Psychiatry journal. Since 2019 there has been ongoing debate around the accuracy of that 56% withdrawal symptom rate, with concerns around the potential for bias, for example, by including data such as online surveys. This new paper aimed to avoid bias by focusing on RCTs and observational studies assessing the incidence of antidepressant discontinuation of an established antidepressant drug against placebo.

79 studies were assessed as appropriate and included (44 RCTs and 35 observational studies) with a total of 21k patients. 

It found around 31% of patients reported at least one antidepressant discontinuation symptom compared to 17% on placebo. Thankfully only 3% of patients on ADs had a severe response, while it very unlikely with placebo (0.6%). If just examining RCTs alone, the difference in rate of withdrawal effects between antidepressant and placebo was only 8%. 

This data firstly confirms that withdrawal effects on antidepressant cessation is a genuine problem, but reassuringly suggests the rate is less than earlier data has reported. 

Secondly, it shows if a patient has severe symptoms it is likely to be a true discontinuation reaction. For these patients ongoing clinician input is very important, with assessment of whether symptoms may be improving and therefore time and support is the primary need, or whether restarting the medication may be required for stabilisation and then discussion about whether further discontinuation attempts are desired and alternative approaches could be more effective. 

Thirdly, this data suggests that patient expectation and how clinicians deliver information about withdrawal effects may have a profound effect on the subsequent negative effects a patient experiences. 1 in 6 patients on placebo experience nocebo withdrawal effects, accounting for at least half the effect experienced by people on antidepressants. 

Does this mean it is better to actively not discuss withdrawal effects when stopping antidepressants? No, I think the discussion remains important – enough people have unpleasant effects that it needs to be addressed upfront, in advance, so patients know their symptoms do not necessarily indicate a relapse of their original mental health issue, and that there a various strategies to help. 

But this data could have a positive effect for patients and help mitigate some of that nocebo response. Knowing that ONLY 1 in 6 people suffer withdrawal symptoms, and only 1 in 35 severe ones, is clear and useful knowledge to share. Concise too. Perhaps it may help me speed up my consultations after all...